UvA-DARE is a service provided by the library of the University of Amsterdam (http
s
://dare.uva.nl)
UvA-DARE (Digital Academic Repository)
Affective Teacher–Student Relationships and Students' Engagement and
Achievement
A Meta-Analytic Update and Test of the Mediating Role of Engagement
Roorda, D.L.; Jak, S.; Zee, M.; Oort, F.J.; Koomen, H.M.Y.
DOI
10.17105/SPR-2017-0035.V46-3
Publication date
2017
Document Version
Final published version
Published in
School Psychology Review
Link to publication
Citation for published version (APA):
Roorda, D. L., Jak, S., Zee, M., Oort, F. J., & Koomen, H. M. Y. (2017). Affective
Teacher–Student Relationships and Students' Engagement and Achievement: A Meta-
Analytic Update and Test of the Mediating Role of Engagement.
School Psychology Review
,
46
(3), 239-261 . https://doi.org/10.17105/SPR-2017-0035.V46-3
General rights
It is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s)
and/or copyright holder(s), other than for strictly personal, individual use, unless the work is under an open
content license (like Creative Commons).
Disclaimer/Complaints regulations
If you believe that digital publication of certain material infringes any of your rights or (privacy) interests, please
let the Library know, stating your reasons. In case of a legitimate complaint, the Library will make the material
inaccessible and/or remove it from the website. Please Ask the Library: https://uba.uva.nl/en/contact, or a letter
to: Library of the University of Amsterdam, Secretariat, Singel 425, 1012 WP Amsterdam, The Netherlands. You
will be contacted as soon as possible.
Download date:09 Aug 2022
239
School Psychology Review
2017, Volume 46, No. 3, pp. 239–261
DOI: 10.17105/SPR-2017- 0035.V46 -3
Affective Teacher–Student Relationships and Students’
Engagement and Achievement: A Meta-Analytic Update
and Test of the Mediating Role of Engagement
Debora L. Roorda
Suzanne Jak
Marjolein Zee
Frans J. Oort
Helma M. Y. Koomen
University of Amsterdam
Abstract. The present study took a meta-analytic approach to investigate whether students’ engagement acts as a
mediator in the association between affective teacher–student relationships and students’ achievement. Furthermore,
we examined whether results differed for primary and secondary school and whether similar results were found in
a longitudinal subsample. Our sample consisted of 189 studies (249,198 students in total) that included students
from preschool to high school. A distinction was made between positive relationship aspects (e.g., closeness) and
negative relationship aspects (e.g., conict). Meta-analytic structural equation modeling showed that, overall, the
associations between both positive relationships and achievement and negative relationships and achievement were
partially mediated by student engagement. Subsequent analyses revealed that mediation is applicable to both
primary and secondary school. Only the direct association between positive relationships and engagement was
stronger in secondary school than in primary school. Finally, partial mediation was also found in the longitudinal
subsample.
It has been repeatedly suggested that the affective quality
of dyadic teacher–student relationships inuences students’
engagement and achievement (e.g., Hamre & Pianta, 2001;
Hughes, 2011). A previous meta-analysis provided support for
these associations (Roorda, Koomen, Spilt, & Oort, 2011). As
the last 5 years have seen an exponential increase in studies
about this topic, we updated the previous meta-analytic sample
with recent studies. Moreover, it has generally been assumed
that the association between teacher–student relationships and
academic achievement can be explained by students’ engage-
ment (Connell & Wellborn, 1991; Tucker et al., 2002).
However, relatively few empirical studies have actually exam-
ined this mediating role of students’ engagement, and available
studies showed somewhat inconsistent results (e.g., Hughes,
Luo, Kwok, & Loyd, 2008; Lam et al., 2012). In the present
study, meta-analytic structural equation modeling (MASEM)
was used, which enabled us to investigate the mediating role of
engagement in a meta-analytic sample with studies that often
did not test this mediating role themselves. Thus, the present
study contributes to the eld in several ways: First, it provides
an update to a previous meta-analysis on teacher–student rela-
tionships, engagement, and achievement. Second, it allows
conclusions about mediation by engagement based on a much
larger number of studies than before. Third, because there are
differences between primary and secondary school, both in the
professional roles of teachers (Bergin & Bergin, 2009;
Hargreaves, 2000) and methodological characteristics of stud-
ies (Roorda et al., 2011), we examine whether engagement
similarly explains associations between affective teacher–stu-
dent relationships and achievement across primary and second-
ary school. Finally, as our meta-analysis also includes a
substantial number of longitudinal studies, we are able to inves-
tigate whether direct and indirect associations among teacher–
student relationships, engagement, and achievement hold over
Author Note. This research was supported by grant 411-08-502 from the Netherlands Organization for Scientic Research assigned to Helma
Koomen.
Correspondence regarding this article should be addressed to Debora Roorda, Research Institute of Child Development and Education,
University of Amsterdam, P.O. Box 15776, NL-1001 NG Amsterdam, The Netherlands. Tel.: +3120 525 8198; e-mail: D.L.Roorda@
uva.nl.
Copyright 2017 by the National Association of School Psychologists. ISSN 0279-6015, eISSN 2372-966x
School Psychology Review, 2017, Volume 46, No. 3
240
DOI: 10.17105/SPR-2017- 0035.V46 -3
time (i.e., are also found in a subsample with longitudinal stud-
ies only).
Theoretical Perspectives
Two theoretical approaches that have been especially
important in research about teacher–student relationships in
connection with academic adjustment are social–motivational
theories and the extended attachment perspective (Davis,
2003). Both approaches assume that students’ engagement
plays an important role in explaining the impact of teacher–
student relationships on students’ achievement. Students’
engagement is considered to be a multidimensional concept
and can be dened as “the quality of a student’s connection or
involvement with the endeavor of schooling and hence with
the people, activities, goals, values, and place that compose
it” (Skinner, Kindermann, & Furrer, 2009, p. 494).
Engagement thus includes different aspects that can be orga-
nized in three broad components (Fredricks, Blumenfeld, &
Paris, 2004): behavioral engagement, which refers to stu-
dents’ participation in academics as well as social or extracur-
ricular activities (e.g., effort, persistence, concentration);
emotional engagement, which describes students’ positive
and negative feelings and reactions to academics, teachers,
classmates, and school (e.g., enjoyment, satisfaction, bore-
dom); and cognitive engagement, which refers to students’
thoughtfulness and willingness to invest in the mastering of
difcult skills and comprehension of complex ideas (e.g.,
self-regulation, cognitive strategy use). In most empirical
studies, however, either these three components are not clearly
distinguished or only one or two aspects of behavioral and/or
emotional engagement are investigated. Therefore, we made
no subdivisions and examined engagement as one multidi-
mensional concept.
According to social–motivational theories (Connell &
Wellborn, 1991; Deci, Vallerand, Pelletier, & Ryan, 1991),
students will become engaged in schoolwork if their basic
psychological needs for relatedness, competence, and auton-
omy are met. Teachers can fulll these needs by respectively
showing involvement (caring for and expressing interest in
the student), providing structure (establishing clear rules and
consequences), and supporting autonomy (giving students
freedom to make their own choices). Students’ higher engage-
ment will, in turn, lead to higher grades and better perfor-
mance on achievement tests (Skinner, Wellborn, & Connell,
1990). In this way, teachers’ supportive behaviors affect stu-
dents’ achievement through their impact on students’
engagement.
In the extended attachment perspective (Pianta, 1999;
Verschueren & Koomen, 2012), the mediating role of engage-
ment has been less explicitly hypothesized. A central idea in
attachment theory is that teacher–student relationships that
are characterized by high levels of closeness (i.e., the degree
of warmth and openness in the relationship) and low levels of
conict (i.e., discordant and coercive interactions between
teacher and student; Pianta, 2001) will help children feel
emotionally secure. Emotional security, in turn, enables chil-
dren’s exploration of the learning environment and engage-
ment in academic activities, which will result in better
academic performance (Bergin & Bergin, 2009; Koomen, van
Leeuwen, & van der Leij, 2004). In this way, teacher–student
relationships inuence school achievement through children’s
exploration of the environment and engagement in learning
activities.
In the present meta-analysis, we focused on the affec-
tive quality of teacher–student relationships, or the emo-
tional quality of interactions between teachers and students,
as well as teachers’ and students’ feelings and beliefs about
each other and their mutual relationship (Pianta, Hamre, &
Stuhlman, 2003). We chose this focus because a previous
meta-analysis showed that affective teacher behaviors (e.g.,
teacher empathy and teacher warmth) were more strongly
associated with students’ school outcomes than instruc-
tional teacher behaviors (e.g., encouragement of learning
and higher order thinking; Cornelius-White, 2007).
Likewise, previous studies found evidence that the affective
dimension of social–motivational theory, teacher involve-
ment, was a more salient predictor of students’ engagement
than teacher structure and autonomy support (e.g., Skinner
& Belmont, 1993; Tucker et al., 2002). Furthermore, in the
present study, we distinguished between positive relation-
ships (e.g., closeness, involvement, relatedness, emotional
support, warmth, and acceptance) and negative relation-
ships (e.g., conict, rejection, role strain, verbal abuse, and
relational negativity). We made this distinction because
studies conducted in primary schools have frequently
reported that negative relationships (i.e., conict) are more
strongly associated with students’ engagement and achieve-
ment than positive relationships (i.e., closeness; Baker,
2006; Hamre & Pianta, 2001).
Empirical Support for the Explanatory Role of
Engagement
Although the mediating role of engagement has been
generally assumed, relatively few empirical studies have actu-
ally investigated whether students’ engagement acts as a
mediator in the association between teacher–student relation-
ships and students’ academic achievement. Concerning the
mediating role of engagement for positive relationships,
Woolley, Kol, and Bowen (2009) showed that the association
between teacher support and students’ grades was mediated
by students’ satisfaction with school in a sample of Latino
middle school students. Similarly, Zimmer-Gembeck,
Chipuer, Hanisch, Creed, and McGregor (2006) reported that
high school students’ engagement mediated the link between
teacher–student relationships and self-reported grades. In a
sample of third-grade students, O’Connor and McCartney
(2007) found that the association between teacher–student
relationship quality and achievement test scores was only par-
tially mediated by students’ classroom engagement. Likewise,
Lam et al. (2012) found that engagement partially mediated
Mediating Role of Engagement
241
the association between teacher support and performance in
a sample of seventh to ninth graders from 12 different coun-
tries. With regard to negative relationships, de Bruyn (2005)
found evidence that students’ attentiveness fully mediated the
link between teacher role strain (measured with items such as
“Teachers don’t seem to like me” and “Many teachers don’t
know me”) and grade point average in a sample of rst-year
secondary school students.
Incidental longitudinal studies have also found evi-
dence for the mediating role of engagement. In a 3-year lon-
gitudinal study with primary school students who were
academically at risk due to their low literacy levels, Hughes
et al. (2008) revealed that the association between teacher
perceptions of positive relationship quality in Grade 1 and
children’s scores on both math and reading achievement tests
in Grade 3 was fully mediated by engagement in Grade 2. In
a follow-up study with the same sample, Hughes et al. found
that engagement in Grade 4 also mediated the link between
teacher–student conict in Grade 3 and math and reading test
scores in Grade 5 when students reported about relationship
quality (Hughes, Wu, Kwok, Villarreal, & Johnson, 2012).
Likewise, Kiuru et al. (2014) showed that the association
between positive teacher affect in Grade 1 and academic per-
formance in Grade 4 was fully mediated by task-focused
behavior in Grades 2 and 3 in a sample of Finnish primary
school children.
Although these studies provided evidence for the
hypothesis that engagement mediates the association between
teacher–student relationships and achievement, results are
somewhat inconsistent. Some studies found evidence for full
mediation (e.g., Hughes et al., 2008), whereas others only
provided support for partial mediation (e.g., Lam et al., 2012).
Furthermore, a large portion of the studies that measured
teacher–student relationships, engagement, and achievement
did not test the mediating role of engagement. Therefore,
empirical evidence remains limited to a relatively small num-
ber of studies conducted in specic samples with specic age
groups and measuring different aspects of engagement. In the
present study, we used a large sample of studies (k = 189) to
draw more substantiated conclusions about the mediating role
of engagement.
Previous Meta-Analysis on Teacher–Student
Relationships, Engagement, and Achievement
Results of a previous meta-analysis on affective
teacher–student relationships, engagement, and achievement
(Roorda et al., 2011) also gave rise to the assumption that
engagement plays a mediating role; associations between
teacher–student relationships and engagement were stronger
than associations between teacher–student relationships and
achievement (Roorda et al., 2011). The smaller effect sizes for
teacher–student relationships and achievement seemed to sug-
gest that associations were partly indirect (i.e., mediated by
engagement). In the present study, we updated our meta-ana-
lytic sample and used more advanced statistical techniques
(MASEM; Cheung & Chan, 2005; Jak, 2015; Jak, Oort,
Roorda, & Koomen, 2013; Viswesvaran & Ones, 1995) to
examine whether the smaller effect sizes for achievement
were actually due to an indirect effect of teacher–student rela-
tionships through engagement. An advantage of this new tech-
nique is that indirect effects can be tested on a meta-analytic
level, even if the original studies do not provide statistical
information about indirect effects (Becker, 1992; Viswesvaran
& Ones, 1995).
Differences Between Primary and Secondary School
There are considerable differences between primary
school and secondary school. First, secondary school students
usually have several teachers during the school day, whereas
primary school students generally spend most of their time
with the same teacher. Additionally, secondary schools are
mostly larger than primary schools. Perhaps due to this scale
difference, teacher–student relationships are usually less per-
sonal, less positive, and more distant in secondary schools
than in primary schools (Bergin & Bergin, 2009; Hargreaves,
2000). These differences (i.e., less contact moments and more
distant relationships between teachers and students) could
imply that relationships with teachers are less important for
the engagement and achievement of secondary school stu-
dents than primary school children. This line of reasoning is
supported by the general assumption made in the literature
that secondary school students become increasingly indepen-
dent from teachers and more focused on peers (e.g.,
Buhrmester & Furman, 1987; Hargreaves, 2000; Lynch &
Cicchetti, 1997). However, it could also be argued that fewer
contact moments and less positive bonds between teachers
and students make secondary school students more sensitive
to the degree of warmth and support they receive from their
teachers, and hence, teacher–student relationships might be
more important for secondary school students’ engagement
and achievement (cf., Roorda et al., 2011). Therefore, it is
important to investigate the mediation models separately for
primary and secondary school studies. Moreover, there are
also methodological differences between primary and second-
ary school studies. Secondary school studies more often use
student reports for both teacher–student relationships and
engagement, as well as a cross-sectional design, and therefore
some associations might be stronger in secondary school stud-
ies (Roorda et al., 2011).
The Present Study
In the present study, we rst updated the literature search
of a previous meta-analysis (Roorda et al., 2011). In this way,
we were able to investigate whether associations among
teacher–student relationships, engagement, and achievement
would still hold if studies from the past 5 years were included.
Second, we investigated whether students’ engagement medi-
ated the association between affective teacher–student rela-
tionships and academic achievement in the total sample of
School Psychology Review, 2017, Volume 46, No. 3
242
DOI: 10.17105/SPR-2017- 0035.V46 -3
studies. Third, we examined whether the mediational model
applied to both primary and secondary school studies. Fourth,
although most studies in our meta-analytic sample used a
cross-sectional design, there was also a substantial number of
longitudinal studies (k = 52), and we were therefore able to
investigate whether the mediating role of engagement was
found over time (i.e., in the longitudinal subsample only).
Figure 1a provides an overview of the hypothesized
model. With regard to the direct effects, we hypothesized that
positive relationships would be positively associated with stu-
dents’ engagement (β
31
in the model; e.g., Murray & Zvoch,
2011; Tucker et al., 2002), whereas negative relationships
would be negatively linked to engagement (β
32
; e.g., de Laet
et al., 2015; Murray, 2009). Furthermore, we expected that
students’ engagement would be positively associated with
students’ achievement (β
43
; e.g., Chen & Gregory, 2009; You,
Hong, & Ho, 2011). In addition, if direct effects could be
identied, we hypothesized we would nd positive associa-
tions between positive relationships and achievement (β
41
;
e.g., Murray & Zvoch, 2011; You et al., 2011) and negative
associations between negative relationships and achievement
(β
42
; e.g., Al-Yagon & Mikulincer, 2004; Pianta, Nimetz, &
Bennett, 1997). Finally, we expected to nd indirect effects
from both positive relationships through engagement on
achievement (β
31
× β
43
; Hughes et al., 2012; Kiuru et al.,
2014) and from negative relationships through engagement
on achievement (β
32
× β
43
; de Bruyn, 2005). To summarize,
we expected that engagement would mediate the association
between both positive and negative relationships and students’
achievement in the total sample of studies. However, as some
previous studies found evidence for full mediation (e.g.,
Hughes et al., 2008) and others for partial mediation (e.g.,
Lam et al., 2012), we were not sure which to expect.
In the previous meta-analysis, associations between
teacher–student relationships and engagement were stronger
than associations between teacher–student relationships and
achievement both in primary and secondary school studies.
Therefore, we hypothesized that the mediating role of engage-
ment would be applicable to both primary and secondary
school. Still, associations between positive relationships and
both engagement and achievement were stronger in secondary
school studies, whereas associations between negative relation-
ships and both engagement and achievement were stronger in
primary school studies (Roorda et al., 2011). Therefore, asso-
ciations with positive relationships were expected to be stronger
in secondary school, whereas associations with negative rela-
tionships were anticipated to be stronger in primary school.
Finally, we expected that the mediating role of engagement
would also be found in the longitudinal subsample (e.g.,
Hughes et al., 2008; Kiuru et al., 2014) but that associations
would probably be weaker than in the total sample of studies.
METHOD
In the present study, we used data from a previous
meta-analysis (Roorda et al., 2011). In addition, we performed
a new literature search to nd studies that were published
between 2009 and 2016. A description of this literature search
Figure 1. (a) Mediation Model With Parameter Labels and (b) Results for the Final Model for the
Total Sample of Studies (k = 189)
